1. 增加api的一键启动脚本;2. 增加webui的一键启动脚本;3. 更新readme;4. 调整llm_api_launch

README.md

@@ -257,7 +257,7 @@ $ python server/llm_api_launch.py --model-path-addresss model1@host1@port1 model
 $ python server/llm_api_launch.py --gpus 0,1 --num-gpus 2 --max-gpu-memory 10GiB
 ```
 
-注：以如上方式启动LLM服务会以nohup命令在后台运行 fastchat 服务，如需停止服务，可以运行如下命令：
+注：以如上方式启动LLM服务会以nohup命令在后台运行 fastchat 服务，如需停止服务，可以运行如下命令,但该脚本**仅适用于linux和mac平台**：
 
 ```shell
 $ python server/llm_api_shutdown.py --serve all 
@@ -331,6 +331,55 @@ $ streamlit run webui.py --server.port 666
 
 ---
 
+### 6 一键启动：
+
+#### 6.1 api服务一键启动脚本
+
+新增api一键启动脚本，可一键开启fastchat后台服务及本项目提供的langchain api服务,调用示例：
+
+调用默认模型：
+
+```shell
+$ python server/api_allinone.py
+```
+
+加载多个非默认模型：
+
+```shell
+$ python server/api_allinone.py --model-path-address model1@host1@port1 model2@host2@port2
+```
+
+多卡启动：
+
+```shell
+python server/api_allinone.py --model-path-address model@host@port --num-gpus 2 --gpus 0,1 --max-gpu-memory 10GiB
+```
+其他参数详见各脚本及fastchat服务说明。
+
+#### 6.2 webui一键启动脚本
+加载本地模型：
+```shell
+$ python webui_allinone.py
+```
+
+调用远程api服务：
+```shell
+$ python webui_allinone.py --use-remote-api
+```
+后台运行webui服务：
+```shell
+$ python webui_allinone.py --nohup
+```
+加载多个非默认模型：
+```shell
+$ python webui_allinone.py --model-path-address model1@host1@port1 model2@host2@port2 
+```
+多卡启动：
+```
+python webui_alline.py --model-path-address model@host@port --num-gpus 2 --gpus 0,1 --max-gpu-memory 10GiB
+```
+其他参数详见各脚本及fastchat服务说明。
+
 ## 常见问题
 
 参见 [常见问题](docs/FAQ.md)。

server/api_allinone.py

@@ -0,0 +1,50 @@
+"""Usage
+调用默认模型：
+python server/api_allinone.py
+
+加载多个非默认模型：
+python server/api_allinone.py --model-path-address model1@host1@port1 model2@host2@port2 
+
+多卡启动：
+python server/api_allinone.py --model-path-address model@host@port --num-gpus 2 --gpus 0,1 --max-gpu-memory 10GiB
+
+"""
+import sys
+import os
+sys.path.append(os.path.dirname(__file__))
+sys.path.append(os.path.dirname(os.path.dirname(__file__)))
+
+from llm_api_launch import launch_all,parser,controller_args,worker_args,server_args
+from api import create_app
+import uvicorn
+
+parser.add_argument("--api-host", type=str, default="0.0.0.0")
+parser.add_argument("--api-port", type=int, default=7861)
+parser.add_argument("--ssl_keyfile", type=str)
+parser.add_argument("--ssl_certfile", type=str)
+# 初始化消息
+args = parser.parse_args()
+args_dict = vars(args)
+
+api_args = ["api-host","api-port","ssl_keyfile","ssl_certfile"]
+
+def run_api(host, port, **kwargs):
+    app = create_app()
+    if kwargs.get("ssl_keyfile") and kwargs.get("ssl_certfile"):
+        uvicorn.run(app,
+                    host=host,
+                    port=port,
+                    ssl_keyfile=kwargs.get("ssl_keyfile"),
+                    ssl_certfile=kwargs.get("ssl_certfile"),
+                    )
+    else:
+        uvicorn.run(app, host=host, port=port)
+
+if __name__ == "__main__":
+    launch_all(args=args,controller_args=controller_args,worker_args=worker_args,server_args=server_args)
+    run_api(
+            host=args.api_host,
+            port=args.api_port,
+            ssl_keyfile=args.ssl_keyfile,
+            ssl_certfile=args.ssl_certfile,
+            )

server/llm_api_launch.py

@@ -181,7 +181,7 @@ base_launch_sh = "nohup python3 -m fastchat.serve.{0} {1} >{2}/{3}.log 2>&1 &"
 # ! 1 log的文件名，必须与bash_launch_sh一致
 # 2 controller, worker, openai_api_server
 base_check_sh = """while [ `grep -c "Uvicorn running on" {0}/{1}.log` -eq '0' ];do
-                        sleep 1s;
+                        sleep 5s;
                         echo "wait {2} running"
                 done
                 echo '{2} running' """
@@ -211,7 +211,7 @@ def string_args(args, args_list):
     return args_str
 
 
-def launch_worker(item):
+def launch_worker(item,args=args,worker_args=worker_args):
     log_name = item.split("/")[-1].split("\\")[-1].replace("-", "_").replace("@", "_").replace(".", "_")
     # 先分割model-path-address,在传到string_args中分析参数
     args.model_path, args.worker_host, args.worker_port = item.split("@")
@@ -225,7 +225,11 @@ def launch_worker(item):
     subprocess.run(worker_check_sh, shell=True, check=True)
 
 
-def launch_all():
+def launch_all(args=args,
+               controller_args=controller_args,
+               worker_args=worker_args,
+               server_args=server_args
+               ):
     controller_str_args = string_args(args, controller_args)
     controller_sh = base_launch_sh.format("controller", controller_str_args, LOG_PATH, "controller")
     controller_check_sh = base_check_sh.format(LOG_PATH, "controller", "controller")
@@ -233,7 +237,7 @@ def launch_all():
     subprocess.run(controller_check_sh, shell=True, check=True)
 
     if isinstance(args.model_path_address, str):
-        launch_worker(args.model_path_address)
+        launch_worker(args.model_path_address,args=args,worker_args=worker_args)
     else:
         for idx, item in enumerate(args.model_path_address):
             print(f"开始加载第{idx}个模型:{item}")

webui_allinone.py

@@ -0,0 +1,73 @@
+"""Usage
+加载本地模型：
+python webui_allinone.py
+
+调用远程api服务：
+python webui_allinone.py --use-remote-api
+
+后台运行webui服务：
+python webui_allinone.py --nohup
+
+加载多个非默认模型：
+python webui_allinone.py --model-path-address model1@host1@port1 model2@host2@port2 
+
+多卡启动：
+python webui_alline.py --model-path-address model@host@port --num-gpus 2 --gpus 0,1 --max-gpu-memory 10GiB
+
+"""
+import streamlit as st
+from webui_pages.utils import *
+from streamlit_option_menu import option_menu
+from webui_pages import *
+import os
+from server.llm_api_launch import string_args,launch_all,controller_args,worker_args,server_args,LOG_PATH
+
+from server.api_allinone import parser, api_args
+import subprocess
+
+parser.add_argument("--use-remote-api",action="store_true")
+parser.add_argument("--nohup",action="store_true")
+parser.add_argument("--server.port",type=int,default=8501)
+parser.add_argument("--theme.base",type=str,default='"light"')
+parser.add_argument("--theme.primaryColor",type=str,default='"#165dff"')
+parser.add_argument("--theme.secondaryBackgroundColor",type=str,default='"#f5f5f5"')
+parser.add_argument("--theme.textColor",type=str,default='"#000000"')
+web_args = ["server.port","theme.base","theme.primaryColor","theme.secondaryBackgroundColor","theme.textColor"]
+
+args = parser.parse_args()
+
+def launch_api(args=args,args_list=api_args,log_name=None):
+    print("launch api ...")
+    if not log_name:
+        log_name = f"{LOG_PATH}api_{args.api_host}_{args.api_port}"
+    print(f"logs on api are written in {log_name}")
+    args_str = string_args(args,args_list)
+    api_sh = "python  server/{script} {args_str} >{log_name}.log 2>&1 &".format(
+        script="api.py",args_str=args_str,log_name=log_name)
+    subprocess.run(api_sh, shell=True, check=True)
+    print("launch api done!")
+
+
+def launch_webui(args=args,args_list=web_args,log_name=None):
+    print("Launching webui...")
+    if not log_name:
+        log_name = f"{LOG_PATH}webui"
+    print(f"logs on api are written in {log_name}")
+    args_str = string_args(args,args_list)
+    if args.nohup:
+        webui_sh = "streamlit run webui.py {args_str} >{log_name}.log 2>&1 &".format(
+        args_str=args_str,log_name=log_name)
+    else:
+        webui_sh = "streamlit run webui.py {args_str}".format(
+        args_str=args_str)
+    subprocess.run(webui_sh, shell=True, check=True)
+    print("launch webui done!")
+
+
+if __name__ == "__main__":
+    print("Starting webui_allineone.py, it would take a while, please be patient....")
+    if not args.use_remote_api:
+        launch_all(args=args,controller_args=controller_args,worker_args=worker_args,server_args=server_args)
+    launch_api(args=args,args_list=api_args)
+    launch_webui(args=args,args_list=web_args)
+    print("Start webui_allinone.py done!")